Genome-wide Surveillance of Transcription Errors in Eukaryotic Organisms

1. Preparation

1. RNases are omnipresent; therefore, clean the workspace thoroughly by spraying it down with a decontamination reagent (e.g., RNase AWAY) and 70% ethanol. Spray down any pipettes, pens, or tube racks to remove potential sources of RNase contamination.
2. To further prevent RNases from contaminating the samples, wear a lab coat or long-sleeve T-shirt during the experimentation. Avoid contact between the gloves and the inside of any tubes that will be used, especially when retrieving them from a box or bag.
3. Prior to the reverse transcription, change gloves frequently.
   NOTE: For optimal results, do not pause the experiment prior to the reverse transcription.

2. Cell and Animal Culture and Collection

1. Saccharomyces cerevisiae growth and collection
   1. If a C-seq library is desired from yeast, first inoculate a single colony of S. cerevisiae in 3 mL of medium containing yeast extract, adenine, peptone, and dextrose (YAPD), and incubate it overnight at 30 °C in a rolling drum.
   2. In the morning, re-inoculate the culture in 50 mL of YAPD medium to an optical density (OD) of 0.1 and grow the cells until they reach an OD of 0.5–0.7 (~7 x 10⁶ cells/mL).
   3. Collect the entire 50 mL culture (approximately 350 x 10⁶ cells) in a 50 mL conical tube and spin the cells down for 3 min at 2,100 x g. Carefully remove the YAPD medium and wash the pellet once with 1x PBS.
   4. Then, resuspend the pellet in 480 µL of lysis buffer, 48 µL of 10% SDS, and 480 µL of Phenol:Chloroform:Isoamyl alcohol (25:24:1), pH 6.8. Vortex the sample at maximum speed for 5 s, and transfer it to a 1.5 mL screw cap tube with 750 µL of ice-cold zirconia beads (supplied with the kit). Proceed to step 3 of this protocol for the cell lysis and total RNA extraction.
      NOTE: All of the reagents required for step 2.1.4 are provided in the recommended yeast RNA extraction kit. These reagents work equally well for yeast, worms, and flies so that after the collection of the yeast (step 2.1), worms (step 2.2) or flies (step 2.3), a unified approach can be used to generate C-seq libraries (steps 3–10).
2. Caenorhabditis elegans culture and collection
   1. If a C-seq library is desired from worms, deposit 30 adult worms on a fresh plate of agar spotted with OP50 bacteria.
      NOTE: Five plates are sufficient for 1 replicate.
   2. Culture the worms for 4 days and collect the worms by gently washing the plates with 5 mL of M9 media, and pool the worms into a single 50 mL conical tube.
   3. Spin the worms down at 100 x g for 2 min to create a pellet. Decelerate the centrifuge at the lowest speed possible, so as not to disturb the pellet.
   4. Gently move the worms into a 1.5 mL tube, wash them twice in 1.5 mL of M9 media, and centrifuge the tube at 100 x g for 1 min to remove bacteria and generate a clean 100 µL pellet of worms.
   5. Resuspend the worms in 480 µL of lysis buffer, 48 µL of 10% SDS, and 480 µL of Phenol:Chloroform:Isoamyl alcohol. Vortex the sample at maximum speed for 5 s and transfer it to a 1.5 mL screw cap tube containing 750 µL of ice-cold zirconia beads. Proceed to step 3 of this protocol for the lysis of the worms and the total RNA extraction.
3. Drosophila melanogaster culture and collection
   1. If a C-seq library is desired from flies, culture 20–25 flies in vials that contain either dextrose or a molasses-based medium. Aliquot the flies over 3 vials, so that each vial contains approximately 8 flies.
   2. To collect the flies, place the vial into an ice bucket until the flies are sedated. Then, collect them with a brush in a 1.5 mL tube. Do not collect the flies by CO₂ treatment.
   3. Let the flies warm to room temperature (RT), and quickly add a premade mixture of 500 µL of lysis buffer, 50 µL of 10% SDS, and 500 µL of Phenol:Chloroform:Isoamyl alcohol to the tube. Use a micropestle to grind up the flies with approximately 15 strokes, accompanied by a firm twisting motion when the pestle reaches the bottom of the tube.
   4. Pour the mixture of flies and lysis media into a 1.5 mL screw cap tube containing 750 µL of ice-cold zirconia beads, and vortex it at maximum speed for 5 s. Proceed to step 3 of this protocol for the lysis of the flies and the total RNA extraction.

3. Total RNA Purification

NOTE: At this point, all three protocols converge, and a single, unified approach can be used to generate C-seq libraries.

1. Screw the cap on tightly and attach the tube to a vortexer with an adapter or sticky tape. Vortex the tube at maximum speed for 10 min to lyse the samples. Then, centrifuge the sample at 16,100 x g for 5 min to separate the organic solvents from the aqueous phase.
2. Carefully remove 400–500 µL of the aqueous phase without disturbing the white, cloudy interphase, and add it to a 15 mL conical tube containing 1.9 mL of binding buffer. Mix them thoroughly by vortexing.
3. Add 1.25 mL of 100% ethanol and mix the sample by vortexing. Transfer 700 µL of the sample to a filter cartridge assembled in a collection tube and spin the sample at 14,500 x g for 30 s. Repeat until all of the sample is passed through the cartridge.
   NOTE: At this point, the sample may turn slightly opaque. If too many cells, worms, or flies were lysed, precipitates could appear that may clog up the filter.
4. Wash the filter once with 700 µL of wash solution 1, and twice with 500 µL of wash solution 2 or 3. Finish the washes by spinning the samples at 14,500 x g for 2 min to remove any excess ethanol.
5. Transfer the filter cartridge to a 1.5 mL dolphin tube and add 108 µL of prewarmed elution solution (65 °C).
   NOTE: Never expose RNA samples to temperatures higher than 65 °C prior to a reverse transcription, as doing so will provoke cytosine to uracil deamination events that are impossible to distinguish from transcription errors.
6. Degrade the DNA contamination in the isolated RNA (see steps 2.1–3.4) by adding 11 µL of 10x DNase I buffer, 2 µL of RNase inhibitor, and 4 µL of DNase I (8 U), and incubate them at 37 °C for 30 min.
7. After digestion, add 11 µL of DNase inactivation reagent. Vortex the mixture and let it sit at RT for 5 min. Then, centrifuge the sample at 15,000 x g for 3 min to pellet the inactivation reagent and collect 110 µL of the supernatant without disturbing the pellet. Transfer the supernatant to a 1.5 mL tube.
   NOTE: The DNase inactivation reagent can be quite difficult to pipet properly, so visually inspect the pipette tip after each pipetting action. If the inactivation reagent becomes too viscous to pipet, add 10 mM Tris-HCl (pH 7.4) equal to 1/5 of the remaining volume.
8. (Quality check) Measure the concentration of the total RNA using a spectrophotometer. Then set aside 1 µL of total RNA and dilute it to a concentration of 10 ng/µL in nuclease-free (NF) water. Run the RNA on an appropriate nucleotide analysis instrument with a High Sensitivity RNA Tape to ensure that the RNA is not degraded and has an eRIN value of at least 8.5 or higher (Figure 2B).

4. mRNA Enrichment

1. Perform the mRNA enrichment with an mRNA miniprep kit (see Table of Materials). Add 140 µL of NF water to the sample for a total volume of 250 µL. Then, add 250 µL of 2x binding buffer, mix the sample thoroughly by vortexing, and add 15 µL of oligo(dT)beads.
2. Mix the sample by pipetting it up and down and incubate it at 65 °C for 2 min. Then, place the sample at RT for 10 min. Finally, centrifuge the sample at 16,100 x g for 2 min to pellet the bead:mRNA complexes.
3. Discard the supernatant and resuspend the pellet in 500 µL of wash solution. Transfer the solution to a spin column and spin it for 1 min at 15,000 x g. Discard the flow-through and wash the sample with an additional 500 µL of wash solution. Then, spin the sample for 2 min at 15,000 x g to remove any excess ethanol from the spin filters.
4. Transfer the spin filter to a dolphin tube and resuspend the pellet in the column with 80 µL of prewarmed elution solution (65 °C). Incubate the sample for 1.5 min at 65 °C and elute it by spinning it for 1 min at 15,000 x g.
5. Measure the concentration of purified RNA using a spectrophotometer or similar tool that allows for the quantification of RNA concentration and quality.

5. RNase III Fragmentation and RNA Clean Up

NOTE: (Important) To prepare circular RNA molecules appropriate for generating C-seq libraries, the RNA must be fragmented to roughly 60–80 bases in length. While previous methods have used a chemical fragmentation to fragment RNA, chemical fragmentation with heavy metals introduces damages to the RNA samples that can be misinterpreted as transcription errors during the final analysis. To circumvent this problem, rely instead on a fragmentation using RNase III to generate small fragments. An additional advantage of this enzymatic approach is that it creates compatible ends required for ligation, obviating the need for an end-repair after the chemical fragmentation.

1. Set up the RNA fragmentation reaction with 1 µg of purified RNA, 10 µL of reaction buffer, 5 µL of RNase III, and enough NF water to bring the volume up to 100 µL (see Table of Materials). Add the enzyme last and pipette the mixture up and down at least 10 times to mix. Incubate the sample at 37 °C for 25 min, which tends to produce RNA fragments of approximately 60 – 80 bases in length.
2. Once the 25 min incubation is complete, clean up the reaction immediately with an oligo clean-up and concentrator kit (see Table of Materials) to prevent any further digestion. Add 200 µL of oligo-binding buffer to the sample, mix it thoroughly by vortexing, and add 800 µL of 100% ethanol. Mix the sample by vortexing and transfer it to a provided spin column in a collection tube.
3. Centrifuge the sample at 10,000 x g for 30 s and wash it with 750 µL of wash buffer to remove RNase III. Centrifuge the sample at 10,000 x g for 30 s, discard the flow-through, and wash the sample once more with 750 µL of wash buffer as described above. After the second flow-through has been discarded, centrifuge the column at 15,000 x g for 2 min to remove any excess ethanol.
4. Transfer the column to a fresh 1.5 mL tube and elute the sample in 24 µL of NF water by spinning it at 10,000 x g for 1 min.
5. QUALITY CHECK: Take 2 µL of the purified RNA fragments and dilute it 2-fold by adding 2 µL of NF water to it. Run the fragmented RNA on a screen tape to ensure that the fragmented RNA is in the range of 50–100 bases in length (Figure 2c).

6. RNA Circularization and Rolling Circle Reverse Transcription

1. To circularize the RNA fragments, heat-denature 20 µL of the sample at 65 °C for 1 min and place it on ice immediately for 2 min. Then, add 4 µL of 10x T4 RNA ligase buffer, 4 µL of 10 mM ATP, 1 µL of RNase inhibitor, 1 µL of NF water, and 8 µL of 50% polyethylene glycol (PEG).
2. Mix the sample thoroughly by vortexing, then add 2 µL of 10 U/µL of T4 RNA Ligase I. Mix the sample again by pipetting it and incubate it at 25 °C for 2 h in a thermocycler with the temperature of the lid set at 30 °C.
3. After 2 h, add 10 µL of NF water to the sample and clean up the sample with an oligo clean-up and concentration kit according to the manufacturer’s specifications. Elute the sample in 20 µL of NF water.
4. To reverse transcribe the circular RNA molecules, add 4 µL of 10 mM dNTPs, 4 µL of 50 ng/µL random hexamers, and 9 µL of NF water to 9 µL of RNA. Mix everything thoroughly by pipetting, denature the sample at 65 °C for 1 min, and place it on ice for 2 min. Store the remaining RNA as back-up.
5. Add 8 µL of 5x first-strand synthesis buffer, 2 µL of 0.1 mM DTT, and 4 µL of 200 U/µL reverse transcriptase and mix everything by pipetting. Incubate the sample at 25 °C for 10 min, followed by a 20 min incubation at 42 °C with the lid set at 5 °C higher than the incubation temperature.
   NOTE: A reverse transcriptase with strand displacement capabilities must be used at this step to allow for a rolling circle reverse transcription.
6. Add 10 µL of NF water to the sample and clean it up with the oligo clean-up and concentration kit according to the manufacturer’s recommendations. Elute the sample in 42 µL of elution solution.
7. QUALITY CHECK: Dilute 2 µL of the single-stranded cDNA in 2 µL of NF water and run this sample on a nucleotide analysis instrument with a High Sensitivity RNA Tape. If the reverse transcription worked properly, a library of cDNA molecules, ranging from roughly 60–1,500 bases in length, should be visible. Ideally, most cDNA is in the range of 500–1,000 bases (Figure 2d).

7. Second Strand cDNA Synthesis and End Repair

1. Use a second strand synthesis kit to generate a double-stranded cDNA library. Place the samples on ice and add 30 µL of NF water to 38 µL of your sample. Then, add 8 µL of 10x Second Strand Buffer and 4 µL of Second Strand Enzyme, and mix the sample by gentle pipetting before its incubation at 16 °C for 2.5 h.
2. Clean up the samples with the oligo clean-up and concentration kit according to the manufacturer’s recommendations for 100 µL reactions and elute the samples in 38 µL of NF water.
3. Set up the end-repair reaction by adding 20 µL of NF water to 35.5 µL of double-stranded cDNA, 6.5 µL of End Repair Reaction Buffer and 3 µL of End Prep Enzyme Mix. Carefully check the reaction buffer for white precipitates and dissolve them by hand-warming if present.
4. Mix the end-repair reaction by pipetting and incubate it at 20 °C for 30 min, followed by a second 30 min incubation at 65 °C. Set the temperature of the thermocycler lid at 5 °C higher than the incubation temperature.

8. Adapter Ligation and Size Selection of Prepared Libraries

1. Use a multi-step library preparation module for the final library preparations (see Table of Materials). First, dilute the adaptors for next-generation sequencing 10-fold in 10 mM Tris-HCl to a final concentration of 1.5 µM. Then, add 2.5 µL of diluted adapter and 1 µL of ligation enhancer to each sample and mix them by vortexing.
2. Add 15 µL of Blunt/TA Ligase Master Mix, mix it by pipetting and incubate it at 20 °C for 15 min. Then, add 3 µL of uracil-specific excision reagent enzyme and incubate it at 37 °C for 15 min. After the ligation is complete, add 13.5 µL of NF water to the sample for a total volume of 100 µL and proceed to the size selection.
   NOTE: It is best to size-select for cDNA molecules that are in the range of 600–800 bases in length. If the fragmented RNA molecules were approximately 60–80 bases in length, selecting for molecules that are 600–800 bases in length will ensure that most of the selected molecules contain at least three tandem repeats, which is ideal for the downstream processing and analysis.
3. To select for molecules that are 600–800 bp in length, add 30 µL of resuspended magnetic beads (see Table of Materials), acclimated to RT, to each sample and mix by pipetting. Transfer the sample to a 1.5 mL tube and incubate it at RT for 5 min.
4. Place the sample on a magnetic stand for 5 min to separate the beads from the supernatant. Transfer the supernatant (which contains the desired cDNA molecules) to a new 1.5 mL tube and dispose of the magnetic beads (which are bound to large cDNA molecules).
5. Add a fresh aliquot of 15 µL of magnetic beads to each samples, mix thoroughly by pipetting, and incubate for 5 min at RT. Place the samples on a magnetic rack and incubate them for 5 min at RT. Then, carefully remove and dispose of the supernatants, making sure not to disturb the beads.
   NOTE: Do not dispose of the magnetic beads, which now contain the desired target cDNA.
6. Add 200 µL of 80% freshly prepared ethanol to each of the samples without disturbing the pelleted magnetic beads and incubate for 30 s. Discard the ethanol and wash the samples once more with 80% ethanol. Then, fully remove any trace of ethanol from the samples and air-dry the beads for 5 min but be careful not to over-dry the beads. If the beads are overdried, cracks will appear in the pellets.
7. To elute the samples from the beads, remove the tubes from the magnetic stand and add 19 µL of 10 mM Tris-HCl. Pipet it up and down multiple times to resuspended the beads and incubate the samples for 5 min at RT. If the beads were overdried, incubate them for >10 min.
8. Place the samples back on the magnetic stand and incubate them for 5 min to separate the beads from the supernatant. Carefully remove 15 µL of the supernatant containing the purified, size-selected cDNA libraries from the tubes and transfer it to a fresh 1.5 mL tube.

9. PCR Amplification and Final Bead Purification

1. Add 5 µL of universal primer and 5 µL of a unique index primer (see Table of Materials) to each purified cDNA library and mix them thoroughly by pipetting. Carefully check the polymerase master mix for crystals and other precipitates, and warm the master mix by hand until the precipitates dissolve.
   1. Add 25 µL of the polymerase master mix to the sample, mix them by pipetting, and follow the cycling conditions below for the PCR amplification. Run the initial denaturation at 98 °C for 30 s, and then perform a dual step PCR reaction by denaturating the samples at 98 °C for 10 s and an annealing/extension of the PCR products at 65 °C for 75 s (12x), followed by a final extension at 65 °C for 5 min and an indefinite hold at 4 °C.
2. Clean up the final libraries by adding 45 µL of resuspended magnetic beads directly to the PCR reaction, mix them by pipetting, and incubate them at RT for 5 min. Transfer the sample to a 1.5 mL tube and place it in a magnetic stand. Incubate it for 5 min, discard the supernatant and wash it twice with freshly prepared 80% ethanol for 30 s.
3. After the second wash, fully remove the ethanol from the sample, and air-dry the bead pellet for 5 min and elute it in 35 µL of 0.1x TE. Resuspend the beads by pipetting them and incubate them at RT for 5 min. Place the sample on the magnetic stand and after 5 min, transfer 30 µL of the supernatant to a fresh 1.5 mL tube. Store the libraries at -80 °C.
4. QUALITY CHECK: Dilute 1 µL of the final library in 9 µL of NF water and run the sample on a dedicated DNA analysis instrument with a high-sensitivity chip; the majority of cDNA should be in the range of 600–800 bp in length (Figure 2e).
5. After the elution, send the samples off for sequencing on a massively parallel sequencing platform using 250 bp paired-end reads, or 300 bp single-end reads.

10. Bio-informatic Analysis of Circle Sequencing Data

NOTE: Analyzing and interpreting raw data from the C-seq assay requires a dedicated bio-informatic pipeline. A schematic of the pipeline that was used for our analyses is depicted in Figure 3. Download this pipeline at https://github.com/LynchLab/MAPGD.

1. First, trim the reads to remove adapter sequences and low-quality base calls with cutadapt¹⁸, using a quality score threshold of 20.
2. Then, identify the repeats encoded by the concatemerized RNA molecule using an appropriate algorithm to detect any repeats with a minimum size of 30 nucleotides and a maximum of 2 mismatches between repeats¹⁸. Derive a consensus sequence from these repeats and keep track of all the base calls and associated quality scores at each position.
   NOTE: Because reverse transcription can start at any position on the circularized RNA molecule, the start of a repeat does not necessarily correspond to the 5′-end of the fragmented RNA molecule (hereafter referred to as the ligation point). Accordingly, the consensus sequence will not map continuously against the corresponding transcriptome, which necessitates the identification of the ligation point.
3. To identify the ligation point, map a concatemer of the consensus sequence against the reference transcriptome using the BLAST-like alignment tool (BLAT)¹⁹.
   1. Concatenate two instances of the consensus sequence to ensure that the mapped sequence contains at least one full uninterrupted occurrence of the initial RNA fragment sequence. Then, rotate the consensus sequence to start at the predicted ligation point position.
4. Map the rotated consensus sequences against the genome using TopHat²⁰ and use the alignments to detect any single nucleotide polymorphisms (SNPs) that could be mistaken for transcription errors.
5. Call the transcription errors from the mapped rotated consensus sequences and test whether the candidate errors pass four additional tests.
   1. First, ensure that a transcription error does not overlap with an SNP (a typical requirement is that at least 20 reads need to cover a particular base and no more than 5% of the reads may support an alternative base call).
   2. Second, check that the consensus sequence is derived from at least 3 tandem repeats, each of which calls the same base pair; and thirdly, make sure that the sum of the quality scores of that position must be over 100.
6. For a fourth and final check, rotate the consensus sequence in all possible combinations (simulating all possible positions of the ligation point) and map each version against the reference genome using TopHat.
   NOTE: If one of the rotated sequences finds a perfect match in the genome, the corresponding ligation point is now considered the correct one and the transcription error call should be rejected.